Differential expression of serum/plasma proteins in various infectious diseases: Specific or nonspecific signatures

Apart from direct detection of the infecting organisms or biomarker of the pathogen itself, surrogate host markers are also useful for sensitive and early diagnosis of pathogenic infections. Early detection of pathogenic infections, discrimination among closely related diseases with overlapping clinical manifestations, and monitoring of disease progression can be achieved by analyzing blood biomarkers. Therefore, over the last decade large numbers of proteomics studies have been conducted to identify differentially expressed human serum/plasma proteins in different infectious diseases with the intent of discovering novel potential diagnostic/prognostic biomarkers. However, in-depth review of the literature indicates that many reported biomarkers are altered in the same way in multiple infectious diseases, regardless of the type of infection. This might be a consequence of generic acute phase reactions, while the uniquely modulated candidates in different pathogenic infections could be indicators of some specific responses. In this review article, we will provide a comprehensive analysis of differentially expressed serum/plasma proteins in various infectious diseases and categorize the protein markers associated with generic or specific responses. The challenges associated with the discovery, validation, and translational phases of serum/plasma biomarker establishment are also discussed.     

Acute phase proteins as promising biomarkers: Perspectives and limitations for human and veterinary medicine

Acute phase proteins (APPs) are highly conserved plasma proteins that are increasingly secreted by the liver in response to a variety of injuries, independently of their location and cause. APPs favor the systemic regulation of defense, coagulation, proteolysis, and tissue repair. Various APPs have been applied as general diagnostic parameters for a long time. Through proteomic techniques, more and more APPs have been discovered to be differentially altered. Since they are not consistently explainable by a stereotypic hepatic expression of sets of APPs, most of these results have unfortunately been neglected or attributed to the nonspecificity of the acute phase reaction. Moreover, it appears that various extrahepatic tissues are also able to express APPs. These extrahepatic APPs show focally specific roles in tissue homeostasis and repair and are released primarily into interstitial and distal fluids. Since these focal proteins might leak into the circulatory system, mixtures of hepatic and extrahepatic APP species can be expected in blood. Hence, a selective alteration of parts of APPs might be expected. There are several hints on multiple molecular forms and fragments of tissue‐derived APPs. These differences offer the chance for multiple selective determinations. Thus, specific proteoforms might indeed serve as tissue‐specific disease indicators.     

Floating prioritized subset analysis: A powerful method to detect differentially expressed genes

Controlling the false discovery rate (FDR) is a powerful approach to deal with a large number of hypothesis tests, such as in gene expression data analyses and genome-wide association studies. To further boost power, here we propose a floating prioritized subset analysis (floating PSA) that can more effectively use prior knowledge and detect more genes that are differentially expressed. Genes are first allocated into two subsets: a prioritized subset and a non-prioritized subset, according to investigators’ prior biological knowledge. We allow the FDRs of the two subsets to vary freely (to float) but aim to control the overall FDR at a desired level. An algorithm for the floating PSA is developed to detect the largest number of true positives. Theoretical justifications of the algorithm are given, and computer simulation studies show that the method has good statistical properties. We apply this method to detect genes that are differentially expressed between acute lymphoblastic leukemia and acute myeloid leukemia patients. The result shows that our floating PSA identifies 32 more genes (permutation-based FDR=0.0427) than the conventional (fixed) FDR control. Another example is a colon cancer study, and our floating PSA identifies 43 more genes (permutation-based FDR=0.0502). The floating PSA method is to be recommended for the detection of differentially expressed genes, in light of its power, robustness, and ease of implementation.     

Quantitative proteomic analysis of a paired human liver healthy versus carcinoma cell lines with the same genetic background to identify potential hepatocellular carcinoma markers

To comprehensively measure global changes in protein expression associated with human hepatocellular carcinoma (HCC), comparative proteomic analysis of two cell lines derived from the healthy and carcinoma tissue of a same donor respectively was conducted using quantitative amino acid-coded mass tagging /stable isotope labeling with amino acids in cell culture-based LC-MS/MS approach. Among a total of 501 proteins precisely quantified, the expressions of 128 proteins were significantly altered including 70 proteins up-regulated and 58 down-regulated in HCC cells. According to their previously characterized functions, the differentially expressed proteins were found associated with nine functional categories including glycolysis, stress response, cell communication, cell cycle, apoptosis/death, etc. For example, multiple enzymes involving glycolysis pathway were found differentially regulated in HCC cells, illustrating the critical participation of glycolysis in the HCC transformation. The accuracy of certain differentially expressed proteins identified through the amino acid-coded mass tagging-based quantification was validated in the paired cell lines, and later their pathological correlations were examined in multiple clinical pairs of normal versus tumor tissues from HCC specimen by using a variety of biological approaches including Western blotting and in situ immunoassays. These consistencies suggested that multiple proteins such as HSP27, annexin V, glyceraldehyde-3-phosphate dehydrogenase, nucleolin and elongation factor Tu could be the biomarkers candidates for diagnosis of HCC.     

A proteomic investigation of isogenic radiation resistant prostate cancer cell lines

To model the problem of radiation resistance in prostate cancer, cell lines mimicking a clinical course of conventionally fractionated or hypofractionated radiotherapy have been generated. Proteomic analysis of radiation resistant and radiosensitive DU145 prostate cancer cells detected 4410 proteins. Over 400 proteins were differentially expressed across both radiation resistant cell lines and pathway analysis revealed enrichment in epithelial to mesenchymal transition, glycolysis and hypoxia. From the radiation resistant protein candidates, the cell surface protein CD44 was identified in the glycolysis and epithelial to mesenchymal transition pathways and may serve as a potential therapeutic target.     

Protein profiling in pathology: analysis and evaluation of 239 frozen tissue biopsies for diagnosis of B-cell lymphomas

Purpose : We determined the potential value of protein profiling of tissue samples by assessing how precise this approach enables discrimination of B‐cell lymphoma from reactive lymph nodes, and how well the profiles can be used for lymphoma classification. Experimental design : Protein lysates from lymph nodes (n=239) from patients with the diagnosis of reactive hyperplasia (n=44), follicular lymphoma (n=63), diffuse large B‐cell lymphoma (n=43), mantle cell lymphoma (n=47), and chronic lymphocytic leukemia/small lymphocytic B‐cell lymphoma (n=42) were analysed by SELDI‐TOF MS. Data analysis was performed by (i) classification and regression tree‐based analysis and (ii) binary and polytomous logistic regression analysis. Results : After internal validation by the leave‐one‐out principle, both the classification and regression tree and logistic regression classification correctly identified the majority of the malignant (87 and 96%, respectively) and benign cases (73 and 75%, respectively). Classification was less successful since approximately one‐third of the cases of each group were misclassified according to the histological classification. However, an additional mantle cell lymphoma case that was misclassified as chronic lymphocytic leukemia/small lymphocytic B‐cell lymphoma initially was identified based on the protein profile. Conclusions and clinical relevance : SELDI‐TOF MS protein profiling allows for reliable identification of the majority of malignant lymphoma cases; however, further validation and testing robustness in a diagnostic setting is needed.     

Proteomic profiling of yeast- and hyphal-specific responses of Candida albicans to the antifungal agent, HWY-289

Virulence of Candida albicans is attributable to its unique dimorphic transition from nonpathogenic yeast cells to pathogenic hyphal cells. We previously discovered a novel antifungal agent, known as HWY-289. To characterize the mechanism underlying HWY-289 antifungal activity, we performed 2-DE to identify proteins that were differentially expressed during yeast-to-hyphal transition and in response to HWY-289. Twenty-four differentially expressed protein spots were identified in HWY-289-treated yeast. Most differentially expressed proteins were involved in carbohydrate-derived energy metabolism, cellular detoxification, and antioxidant defenses. Two proteins were involved in cell cycle regulation and DNA processing, and both were downregulated by HWY-289, suggesting that this agent might promote cell death by weakening cellular defense systems. HWY-289 inhibited yeast-to-hyphal transition in a dose-dependent manner. 2-DE analysis of hyphae uncovered several proteins that were induced during yeast-to-hyphal transition. Of these, aconitase and phosphatidylinositol transfer protein were downregulated by HWY-289, suggesting that they mediate the antifungal effects of HWY-289. Finally, RT-PCR analysis revealed that HWY-289 induced expression of three RAS-related genes (CcCST20, CaHST7, and CaCPH1) in yeast cells, but suppressed their expression in hyphae. Thus, the antifungal action of HWY-289 may be attributable to its ability to disrupt prohyphal RAS signaling.     

Comparative proteomic analysis of differentially expressed proteins in an in vitro cellular carcinogenesis model of oral squamous cell carcinoma

In vitro cellular model is an important tool to be used to investigate the cellular events related to pathophysiological conditions in humans. We have developed an in vitro cellular carcinogenesis model of oral squamous cell carcinoma (OSCC). In this study, we performed comparative proteomic analysis using 2‐DE and LC‐tandem mass chromatography to separate and identify differentially expressed proteins. Forty‐five proteins were identified, including 24 proteins with decreased expression and 19 proteins with increased expression during carcinogenesis from immortalized oral epithelial cells to squamous cancerous cells. The identified known proteins were classified into three ontologies of cellular component, molecular function, and biological process. Further validation of five identified proteins (ANXA1, ANXA2, CTSB, KRT17, and S100A6) in the cellular carcinogenesis model and cancerous tissues from OSCC patients confirmed the comparative proteomic results. Moreover, Annexin A1 and A2 expression levels correlated with the pathological differentiation grade of cancerous tissues. Thus, this work provides a dynamic protein file of differentially expressed proteins in oral squamous carcinoma cells, which could provide clues to study the mechanisms of OSCC carcinogenesis and possibly be developed as potential biomarkers for clinical diagnosis or prognostic monitoring.     

Comparison of membrane-associated proteins in human cholangiocarcinoma and hepatocellular carcinoma cell lines

Cholangiocarcinoma (CCA) and hepatocellular carcinoma (HCC) occur with relatively high incidence in Thailand. Cell line models, originating from Thai patients, are available for both diseases, including the human bile duct epithelial carcinoma cell line (HuCCA-1) and the HCC cell line HCC-S102. Here, we have prepared subproteomes enriched in membrane proteins or in cytosolic proteins from the HuCCA-1 and the HCC-S102 cell lines. Study of differential protein expression by 2-DE and LC/MS/MS showed 195 proteins expressed in the two cell lines, including both membrane-associated and cytosolic proteins. Eighteen proteins were found in both membrane and cytosolic fractions of HuCCA-1, but not in HCC-S102, while nine proteins were found in both membrane and cytosolic fractions of HCC-S102, but not in HuCCA-1. Ten membrane proteins were found in HuCCA-1 but not in HCC-S102, including integrin alpha-6 precursor, ezrin, hippocalcin-like protein 1, mitogen-activated protein kinase kinase kinase 2 (MAPK/ERK kinase kinase 2), and calgizzarin. Proteins showing increased expression in the membrane fraction of HuCCA-1 were mainly cytoskeletal proteins (40.9%), while proteins showing increased expression in the membrane fraction of HCC-S102 were mainly metabolic proteins (39.4%). The subproteomic approach used here facilitates detection of potential biomarkers undetected by regular proteomic methods.     

Comprehensive profiling of metastasis-related proteins in paired hepatocellular carcinoma cells with different metastasis potentials

Precise and comprehensive identifications of the proteins associated with metastasis are critical for early diagnosis and therapeutic intervention of hepatocellular carcinoma (HCC). Therefore, we investigated the proteomic differences between a pair of HCC cell lines, originating from the same progenitor, with different metastasis potential using amino acid-coded mass tagging-based LC-MS/MS quantitative proteomic approach. Totally the relative abundance of 336 proteins in these cell lines were quantified, in which 121 proteins were upregulated by >30%, and 64 proteins were downregulated by >23% in the cells with high metastasis potential. Further validation studies by Western blotting in a series of HCC cell types with progressively increasing trend of metastasis showed that peroxiredoxin 4, HSP90β and HSP27 were positively correlated with increasing metastasis while prohibitin was negatively correlated with metastasis potential. These validation results were also consistent with that obtained from comparative analysis of clinic tissues samples. Function annotations of differentially expressed HCC proteome suggested that the emergence and development of high metastasis involved the dysregulation of cell migration, cell cycle and membrane traffics. Together our results revealed a much more comprehensive profile than that from 2-DE-based method and provided more global insights into the mechanisms of HCC metastasis and potential markers for clinical diagnosis.     

Modulation of cancer cell line differentiation: A neglected proteomic analysis with potential implications in pathophysiology, diagnosis, prognosis, and therapy of cancer

The recent development of compounds that induce cell differentiation in various types of cancer cells has enabled the molecular mechanisms governing this kind of induced cancer regression to be investigated. Moreover, this approach to investigating the pathophysiology of neoplasia represents a promising experimental model for proteomic analysis of cancer cells. Modulating neoplastic cell differentiation grade may reveal cytodifferentiation-related protein expression changes, and doing so in vitro has the advantage of less biological variation. Hence, this analysis brings attention to molecular targets of the so-called differentiating factors (i.e., retinoids, hybrid polar compounds, tyrosine kinase inhibitors, etc.) as well as proteins that are frequently associated with differentiation/dedifferentiation processes. The in vitro study of these proteins and of their pathogenetic roles in cancer may ultimately result in the discovery of cancer biomarkers with diagnostic, prognostic, and therapeutic applications.     

Analysis of the urine proteome in patients with pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDAC) accounts for over 213?000 deaths worldwide each year, largely due to late diagnosis. One of the risk factors for the development of PDAC is chronic pancreatitis (CP); the intense desmoplastic reaction makes differentiation between the two conditions extremely difficult. In order to identify biomarkers for noninvasive diagnosis, we performed 2-D DIGE analysis of urine samples from healthy individuals and patients with PDAC and CP. Despite considerable intersample heterogeneity, a total of 127 statistically valid (p<0.05), differentially expressed protein spots were detected, 101 of which were identified using MALDI-TOF MS. A number of these, including annexin A2, gelsolin and CD59 have already been associated with PDAC, however, their validation using immunoblotting proved challenging. This is probably due to extensive PTMs and processing thus indicating the need for raising specific antibodies for urinary proteins. Despite this, our study clearly demonstrates that urine is a valid source of noninvasive biomarkers in patients with pancreatic diseases.     

Analysis of protein profiling studies of β‐thalassemia/Hb E disease

A number of studies have used global protein profiling technologies on a range of patient samples to detect proteins that are differentially expressed in β‐thalassemia/Hb E as an aid for understanding the physiopathology of this disease. Seven studies have identified a total of 111 unique, differentially expressed proteins. Seven proteins (prothrombin, alpha‐1‐antichymotrypsin, fibrinogen beta chain, hemoglobin beta, selenium‐binding protein, microtubule‐actin cross‐linking factor and adenomatous polyposis coli protein 2) have been identified in two independent studies, whereas two proteins (carbonic anhydrase 1 and peroxiredoxin‐2) have been identified in three independent studies. Both of these latter two proteins were consistently upregulated in the studies that identified them. Ontological analysis of all differentially regulated proteins identified “response to inorganic substances” as the most significant functional annotation cluster, which is consistent with iron overload being a major pathological consequence of this disease. Despite the range of samples investigated and the relatively small number of studies undertaken, a coherent picture of the mediators of the pathological consequences of β‐thalassemia/Hb E disease is starting to emerge.     

On identifying marker genes from gene expression data in a neural framework through online feature analysis

Many attempts have been made to analyze gene expression data. Typical goals of such analysis include discovery of subclasses, designing predictors/classifiers for diseases, identifying marker genes, and trying to get a deeper understanding of underlying biological process. Success of each of these tasks strongly depends on the features used to solve the problem. The high dimensional nature of expression profiles makes the task very difficult. Consequently, many researchers have used some feature selection criteria to reduce the dimensionality of the problem. These approaches are off‐line in nature, as feature selection is done in a separate phase from the system design phase. These approaches ignore the fact that utility of features depends on both the problem that is solved and the tool that is used to solve the problem. We here propose to use a novel neural scheme that picks up the necessary features on‐line when the system learns the classification task. Because it considers all the features at one go, it does not miss any subtle combination of these features. We demonstrate the effectiveness of our on‐line feature selection (OFS) scheme to distinguish between acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) cancer expression data set. Our scheme could identify only five genes that can produce results as good as or even better than what is reported in the literature on this data set. It identifies an important marker gene that alone has a very good discriminating power. This analysis method is quite general in nature and can be effectively used in other areas of bioinformatics. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 453–467, 2006.     

一种基于计算机的白血病细胞治疗效果数学模型

根据熵值在生物系统中的应用原理,改建一种可用计算机模拟的白血病细胞药物治疗效果的数学模型,并应用于新型诱导分化剂氨基甾体对K562白血病细胞的作用中.结果表明药物作用后细胞系统的熵值增加,良好的药物取决于降低癌基因表达率,升高诱导分化率,以致抑制白血病细胞的增殖.该模型亦可用于其他抗癌药物治疗效果的计算机模拟.     

基于覆盖算法的白血病微阵列判别方法

微阵列数据的一个重要应用就是给疾病样本分类。微阵列数据具有样本数量小、特征数量大的特点。该文提出了一种新的方法。以急性白血病的基因表达数据为对象,经过t统计法适当降维,利用覆盖算法白血病类型进行分类,并同已有算法进行了对比。实验结果证明本算法是有效的。     

Bronchoalveolar lavage fluid proteomic patterns of sulfur mustard-exposed patients

Sulfur mustard is an alkylating agent that reacts with ocular, respiratory, cutaneous, and bone marrow tissues. Main late respiratory complications are chronic obstructive pulmonary disease, bronchiectasis, asthma, and bronchiolitis obliterans. The aim of the present study was to identify differentially expressed proteins in bronchoalveolar lavage (BAL) fluid of control healthy and sulfur mustard-exposed lung disease patients. The BAL protein profile of ten healthy and 30 exposed patients with mild, moderate, and severe conditions (ten males in each group) were separated with 2-D SDS-PAGE and differentially expressed protein spots were successfully identified with MALDI TOF TOF MS. Among the differentially expressed proteins we observed a significant increase in vitamin D binding protein isoforms, haptoglobin isoforms, and fibrinogen especially in exposed moderate and severe lung diseases patients (p<0.01). Moreover, compared with healthy controls, significant decreases was noted in calcyphosine, surfactant protein A, and transthyretin in these patients (p<0.01). Apolipoprotein A1 was detected in all patients' BAL fluid but none of the healthy controls. Furthermore, S100 calcium-binding protein A8 was only detected in BAL fluid of moderate and severe groups. These findings will be useful to improve current methods of monitoring and helps to identify new therapeutic targets for treatment of this complicated illness.     

Artificial neural network model for effective cancer classification using microarray gene expression data

Microarray gene expression profile shall be exploited for the efficient and effective classification of cancers. This is a computationally challenging task because of large quantity of genes and relatively small amount of experiments in gene expression data. The repercussion of this work is to devise a framework of techniques based on supervised machine learning for discrimination of acute lymphoblastic leukemia and acute myeloid leukemia using microarray gene expression profiles. Artificial neural network (ANN) technique was employed for this classification. Moreover, ANN was compared with other five machine learning techniques. These methods were assessed on eight different classification performance measures. This article reports a significant classification accuracy of 98% using ANN with no error in identification of acute lymphoblastic leukemia and only one error in identification of acute myeloid leukemia on tenfold cross-validation and leave-one-out approach. Furthermore, models were validated on independent test data, and all samples were correctly classified.